README.md

metadata

title: GAIA LangGraph Agent
emoji: 🤖
colorFrom: blue
colorTo: green
sdk: gradio
sdk_version: 5.31.0
app_file: app.py
pinned: false
hf_oauth: true

GAIA Benchmark Agent with LangGraph

This project implements an agent using the LangGraph framework to solve the GAIA benchmark tasks provided by the Hugging Face Agents Course (Unit 4).

✅ Uses LangChain + LangGraph
✅ Handles 20 benchmark questions
✅ Submits answers for scoring via Hugging Face leaderboard

How to run

1. Clone this repo and push it as a Hugging Face Space (public).
2. Install the dependencies.
3. Launch the Space and log in to Hugging Face.
4. Click "Run Evaluation & Submit All Answers" to evaluate your agent.

Project files

• main.py — Gradio app runner
• graph_builder.py — Builds the LangGraph pipeline
• llm_node.py — Defines LLM node logic
• formatter_node.py — Cleans up answers for exact-match submission
• api_client.py — Wraps GAIA API interactions

Good luck!